Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models741Deals30d268Policy30d67, 15 pendingSources1,014
Pro Mode — raw scores, one benchmark under the lensofficial card harness
The field · 24 measured
10.881.3

24 models measured, most on the official card harness. Seed 1.8 tops the board at 81.3.

24 measured·Lifecycle Updated Oct 8, 2026Score basis
#◆Vendor$/M in$/M out
Leader — ignites whiteField — fill cools with scoreBars show 10.8–81.3 · ◆ solid = first-party or better · outlined = arm's-length
01Seed 1.8ByteDance81.3$0.25/M$2.0/M$0.014
02Qwen3.5 397B A17BAlibaba70.3$0.60/M$3.6/M$0.030
03Qwen3.5 122B A10BAlibaba69.9$0.40/M$3.2/M$0.026
04Qwen3.5 35B A3BAlibaba69.5$0.25/M$2.0/M$0.016
05Step 3.5 Flash+2 altsStepFun66.9$0.10/M$0.30/M$0.003
06GLM-4.7+4 altsZ.ai66.6$0.60/M$2.2/M$0.021
07DeepSeek-V3.2+6 altsDeepSeek65$0.28/M$0.42/M$0.005
08GPT-5+1 altOpenAI633P$1.3/M$10.0/M$0.089
09Kimi K2.5+2 altsMoonshot62.3$0.45/M$2.3/M$0.022
10Kimi K2 Thinking+3 altsMoonshot62.3$0.60/M$2.5/M$0.025
11Qwen3.5 27BAlibaba62.1$0.30/M$2.4/M$0.022
12Qwen3 Max (Reasoning)Alibaba60.9$1.2/M$6.0/M$0.059
13MiMo V2 Flash+2 altsXiaomi51.2$0.10/M$0.30/M$0.004
14GLM-4.6+1 altZ.ai49.53P$0.57/M$2.2/M$0.028
15DeepSeek-V3.1+2 altsDeepSeek49.2$0.56/M$1.7/M$0.023
16MiniMax M2+1 altMiniMax48.5$0.30/M$1.2/M$0.015
17DeepSeek-V3.2-ExpDeepSeek47.9$0.28/M$0.42/M$0.007
18MiniMax M2.1+2 altsMiniMax47.8$0.30/M$1.2/M$0.016
19Grok 4 FastSpaceXAI45$0.20/M$0.50/M$0.008
20Claude Sonnet 4.5+1 altAnthropic42.43P$3.0/M$15.0/M$0.212
21Gemini 2.5 ProGoogle32.23P$1.3/M$10.0/M$0.175
22Claude Sonnet 4Anthropic29.13P$3.0/M$15.0/M$0.309
23Kimi K2 InstructMoonshot28.83P$0.57/M$2.3/M$0.050
24Grok 3SpaceXAI10.8$4.0/M$20.0/M$1.111
◆ solid = independently verified · aggregator attested · vendor attributed — outlined = cross-reference · source-attributed. Dimmed ◆ = confidence inferred. “+N alts” = alternate disclosures in the drawer with why-demoted. Prices per million tokens; $/point = input price ÷ score. Every row opens its model dossier.

Verification: 24 models scored · 0 independently verified · 9 vendor cross-reference · 15 vendor-reported · 0 with source disagreement. How these tiers are assigned